JP5457284B2 - Discourse breakdown calculation system and discourse breakdown calculation program - Google Patents

Description

  The present invention relates to a natural language processing technique, and more particularly, to a discourse breakdown calculation system and a discourse breakdown calculation program for generating breakdown information about the content of a discourse from the result of analyzing the structure of discourse data obtained by speech recognition processing. It is related to effective technology.

  For example, in a call center operation, a so-called operator or communicator (hereinafter sometimes simply referred to as “operator”) needs to perform a series of operations such as creating a response record and storing it in a response management system after receiving a call. There is. At this time, there is a problem that it takes time for the operator to create a response record depending on his / her memory.

  Also, quality (content, uniformity, completeness, conciseness, etc.) varies according to the skill of each operator. In this way, the quality of the response records varies or the quality is low, so it is not possible to accurately grasp the customer's voice, or from the accumulation of response records, new technologies such as text mining are used. Problems such as inability to obtain knowledge also occur.

  In order to solve such a problem, a solution using a speech recognition technology is being studied. By storing the voice information of the response as text using the speech recognition technology, it is possible to ensure the completeness of the response record. It is also possible to automatically create response records by applying natural language processing technology to the text data and summarizing the contents, reducing the cost of response record creation, uniformity, It is possible to improve the usability of response records by ensuring conciseness.

  As such a technique, for example, in Non-Patent Document 1, in voice conversation at a call center, voice data is converted into text, and rules are recursively obtained by inductive learning from past conversation contents and business daily reports. It describes a technology that automatically generates daily business reports by determining and extracting important points for unknown dialogue contents using rules and performing style conversion.

  On the other hand, in call center operations, there is a demand for so-called supervisors to reduce operator response time (cost reduction) and improve response quality (service improvement). For this purpose, for example, training of operators and maintenance of so-called talk scripts and FAQ (Frequently Asked Questions) (hereinafter sometimes simply referred to as “talk scripts”) to be referred to when the operators respond. And so on.

  However, in general, supervisors are busy and often do not have enough time to consider the above-mentioned measures, and the response situation of operators in call center operations, which is a precondition for consideration It is also difficult to accurately grasp the current situation (such as what kind of expressions and contents the operator is responding to the customer, and the use status of the talk script at the time of reception).

Junji Yano, Kenji Araki, “Performance evaluation of daily report generation method using inductive learning for voice conversation in call center”, Information Processing Society of Japan Research Report 2007-NL-178, March 28, 2007, p. 21-28

  As a means for grasping the current situation of the operator's response situation, for example, it is conceivable to monitor and analyze voice information recording the operator's response contents. However, since it takes a lot of time to analyze all the voice information (log), for example, it is conceivable to use data obtained by converting the voice information used when creating the above-mentioned reception record into text.

  Here, when using data obtained by converting a person's discourse into text by speech recognition (hereinafter sometimes referred to as “discourse data”), the text in the discourse data is usually recorded in spoken language, so it is correct. There are cases where existing Japanese language processing technology based on grammatical Japanese sentences or written language cannot be used well. Moreover, the accuracy of the current speech recognition engine cannot eliminate recognition errors, and there are many noises such as typographical errors and omissions due to erroneous recognition. Therefore, it is difficult to analyze and effectively use discourse data using natural language processing.

  On the other hand, in the technology of Non-Patent Document 1, the entire dialogue document is analyzed based on the rules obtained by the inductive learning process, the important sentences are extracted, the style conversion from spoken language to written language is performed, and the daily business report is automatically generated. be able to. However, since the technique of Non-Patent Document 1 first extracts an important sentence from a dialogue document, a summary of the dialogue document can be obtained. However, accumulation of discourse data can be used for other useful applications, for example, as described above. It is difficult to effectively use it for grasping the response situation of a simple operator.

  As another means for grasping the current situation of the operator's response status, for example, the use status of a talk script or FAQ is analyzed from a log or the like. However, in such an analysis, although it is possible to grasp the access history such as the search history and reference history of the talk script and FAQ, whether or not the operator who accessed them actually used it at the time of reception and how it is used. It is impossible to grasp the contents such as whether or not it has been done, and it is necessary to rely on the subjective evaluation of the operator.

  Therefore, the object of the present invention is strong in the characteristics of discourse data such as spoken words and noise, and based on the result of analyzing the structure of discourse data about the discourse by a speaker such as an operator who responds based on reference information such as a talk script, It is an object of the present invention to provide a discourse breakdown calculation system and a discourse breakdown calculation program for efficiently calculating the breakdown information of a discourse by a speaker including information on the use status of reference information. The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

  Of the inventions disclosed in this application, the outline of typical ones will be briefly described as follows.

  A discourse breakdown calculation system according to a representative embodiment of the present invention is a discourse data in which discourse content is converted into text by speech recognition processing, and an analysis result of the discourse structure in the discourse data. Discourse breakdown that uses discourse semantics, which is information to understand the meaning, as input, and calculates and outputs discourse information of the discourse by the speaker, including the usage status of reference information referenced by the speaker during the discourse A calculation system having the following characteristics.

  That is, in the discourse breakdown calculation system, the discourse semantics is a simplification that indicates information about statements that do not correspond to unnecessary expressions in the discourse data, and blocks that consist of one or more consecutive statements for each speaker for these statements. The discourse breakdown calculation system includes information, reception reference information holding information on one or more of the reference information referred to in a discourse, and a block in the discourse data reduced based on the reduction information Calculating the similarity between the text included in the block and the text of each reference information, and obtaining the reference information most similar to the block and outputting it as similarity information Part.

  Further, based on the similarity information, when the similarity for each block in the simplified discourse data is higher than a predetermined threshold, it is determined that the block is uttered according to the corresponding reference information And a discourse breakdown calculation unit that calculates and outputs discourse breakdown information including information on the ratio of blocks uttered according to the reference information and blocks not uttered according to the reference information for each speaker. To do.

  The present invention can also be applied to a program that causes a computer to function as the discourse breakdown calculation system as described above.

  Among the inventions disclosed in the present application, effects obtained by typical ones will be briefly described as follows.

  According to a typical embodiment of the present invention, the structure of discourse data is analyzed with respect to a discourse by a speaker such as an operator who is strong in the characteristics of discourse data such as spoken words and noise and who responds based on reference information such as a talk script. Based on the result, it becomes possible to efficiently calculate the breakdown information of the discourse by the speaker including the information on the usage state of the reference information with sufficient accuracy in practical use.

It is the block diagram shown about the outline | summary of the structural example of the discourse breakdown calculation system which is one embodiment of this invention. It is the block diagram shown about the outline | summary of the structural example of the discourse structure analysis system in one embodiment of this invention. It is the figure shown about the example of the discourse in one embodiment of this invention, and the concept of discourse semantics. It is the figure shown about the flow of discourse data and the example of a context in one embodiment of this invention. It is the figure shown about the example of the flow analysis rule of the discourse structure analysis system in one embodiment of this invention. It is the flowchart which showed the example of the flow of the flow analysis process in the flow analysis part of the discourse structure analysis system in one embodiment of this invention. It is the figure shown about the example of the specific expression analysis rule of the discourse structure analysis system in one embodiment of this invention. It is the flowchart which showed the example of the flow of the specific expression analysis process in the specific expression analysis part of the discourse structure analysis system in one embodiment of this invention. It is the figure shown about the example of cohesion analysis in the discourse structure analysis system in one embodiment of this invention. It is the flowchart which showed the example of the flow of the cohesiveness analysis process in the cohesiveness analysis part of the discourse structure analysis system in one embodiment of this invention. It is the figure shown about the example of the discourse reduction in one embodiment of this invention. It is the flowchart which showed the example of the flow of the discourse reduction process in the discourse reduction part of the discourse structure analysis system in one embodiment of this invention. It is the figure shown about the example of the similarity calculation in the discourse breakdown calculation system in one embodiment of this invention. (A), (b) is the figure shown about the specific example of the discourse breakdown information in one embodiment of this invention. It is the flowchart which showed the example of the flow of the discourse breakdown calculation process in the similarity calculation part and discourse breakdown calculation part of the discourse breakdown calculation system in one embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted.

  The discourse breakdown calculation system according to an embodiment of the present invention is based on discourse semantics (information related to discourse structure for grasping the semantic content of the entire discourse), which is the analysis result of discourse structure regarding discourse data including spoken words. For example, the content of the conversation by a speaker such as a call center operator is determined to be the most similar to the reference information at the time of reception such as a talk script. This is a system that aggregates and outputs as breakdown information.

  Here, discourse semantics means, for example, the meaning content from character strings and morphemes in discourse data, the flow as described later for discourse data, and cohesiveness (discourse segments), proper expressions, discourse reduction It is the analysis result obtained by analyzing the discourse structure such as. This discourse semantics may be generated from discourse data by using a discourse structure analysis system, which will be described later, or may be generated using a sentence structure analysis system using other natural language processing, etc. . It is also possible to manually create a tag by tagging the discourse data. In the present embodiment, description will be made assuming that the generation is performed using a discourse structure analysis system described later.

  The discourse semantics in this embodiment do not have to be the result of strict analysis of the content and structure of Japanese sentences. For example, the discourse semantics have a level of accuracy that does not impede practical use in creating a response record at a call center. I just need it. In this specification, “discourse” refers to an utterance action by one or more speakers, for example, a speech at a lecture where one person speaks, or an exchange between a call center operator and a customer as described above. , Multi-person conference calls and video conferences, and other multi-person conversations and dialogues.

  FIG. 3 is a diagram showing an example of discourse and the concept of discourse semantics in the present embodiment. The example of FIG. 3 shows an example of a discourse consisting of interaction between a call center operator and a customer. The utterance on the left side is the utterance of the operator, and the right side shows the utterance of the customer. In the present embodiment, one sentence of each speaker is called a “statement”. In addition, a group of consecutive statements for each speaker is referred to as a “block” (even if consecutive statements of the same speaker are separated in time, they become different blocks).

  In FIG. 3, “flow” is information that simply indicates the semantic content of each statement. By tracing this information, the flow of discourse can be grasped. For example, in the operator statement of FIG. 3, the first “Yes.” Indicates that the flow is “greeting”, while the next “Yes.” Indicates that the flow is “confusion”. In this way, depending on the discourse situation (context), the flow may be different even for statements of the same wording.

  In FIG. 3, “proprietary expression” is an expression that individually characterizes the content of the discourse. For example, in the discourse of exchange at the call center in FIG. 3, ID, person name, company name, telephone number, location, etc. Applicable to proper nouns. Further, the “segment” is a segmented set of common (united) contents of the discourse for each block. The information about the above “flow”, “specific expression”, and “segment” is a constituent element of “Discourse Semantics” as the analysis result of the discourse structure.

[System configuration (Discourse structure analysis system)]
FIG. 2 is a block diagram showing an outline of a configuration example of the discourse structure analyzing system according to the embodiment of the present invention. The discourse structure analysis system 2 is constituted by a computer system such as a server or a PC (Personal Computer). , The cohesiveness analysis unit 50, the discourse reduction unit 60, the flow analysis rule 22, and the specific expression analysis rule 42, and the discourse data 100 in which the content of the discourse is converted into text by the speech recognition engine 3. Is input, the discourse structure is analyzed, and the discourse semantics 200 is output.

  Note that the existing voice recognition engine 3 can be used. As described above, the discourse structure analysis system 2 according to the present embodiment is characterized by being resistant to noise (such as recognition errors by the speech recognition engine 3) of the discourse data 100, but the recognition rate of the speech recognition engine 3 is higher. Needless to say, it is possible to analyze the discourse structure with higher accuracy.

  The preprocessing unit 10 of the discourse structure analysis system 2 has a function of receiving input of the discourse data 100 and performing preprocessing such as data layout conversion so that processing can be performed in other units as necessary. . Here, for example, continuous statements for each speaker are grouped into blocks, sequence numbers (IDs) indicating the order are assigned to each block and each statement, and speaker data of each block is extracted and set. Perform the process. The discourse data 100 is represented as a data structure using, for example, XML (eXtended Markup Language).

  The flow analysis unit 20 has a function of analyzing a rule-based flow based on the flow analysis rule 22 on the discourse data 100 after the preprocessing by the preprocessing unit 10 and outputting the flow information 21 as an analysis result. . The flow information 21 holds the ID of each statement in the discourse data 100 and the flow information corresponding thereto, and is represented as a data structure using XML or the like, for example.

  This flow information 21 is included in the discourse semantics 200 as an output of the discourse structure analysis system 2 and is an essential component. That is, the discourse semantics 200 includes flow information 21 as a flow analysis result for at least the discourse data 100. Details of the flow analysis rule 22 and details of processing in the flow analysis unit 20 will be described later.

  The morpheme analysis unit 30 performs a so-called morpheme analysis on the discourse data 100 after the preprocessing by the preprocessing unit 10 and divides the discourse data into columns of morphemes (the smallest unit having meaning in the language) to determine the part of speech. As a result, the morpheme information 31 is output. The morpheme information 31 holds the ID of each statement in the discourse data 100 and morpheme string information corresponding thereto, and is represented as a data structure using, for example, XML.

  Similar to the flow information 21, the morpheme information 31 is included in the discourse semantics 200 as an output of the discourse structure analysis system 2, and becomes an input to the specific expression analysis unit 40 and the cohesiveness analysis unit 50 described later. In other words, when the specific expression analysis by the specific expression analysis unit 40 and the cohesiveness analysis (segmentation) by the cohesiveness analysis unit 50 are not performed, the morpheme analysis by the morpheme analysis unit 30 may be unnecessary. In the present embodiment, since the morpheme information 31 is used in the discourse breakdown calculation system described later, the morpheme analysis by the morpheme analysis unit 30 is essential. Note that an existing morpheme analysis engine (for example, MeCab (Japanese cloth candy) or ChaSen (tea candy)) can be used for the morpheme analysis unit 30.

  The specific expression analysis unit 40 is a rule-based specific expression based on the specific expression analysis rule 42 based on the discourse data 100 after the preprocessing by the preprocessing unit 10 and the morpheme information 31 output from the morpheme analysis unit 30. And the specific expression information 41 is output as an analysis result. That is, the analysis process in the specific expression analysis unit 40 is based on the premise that at least the morpheme analysis by the morpheme analysis unit 30 is performed. The specific expression information 41 holds information of each statement ID in the discourse data 100 and a list of specific expressions corresponding to the statement ID, and is represented as a data structure using, for example, XML.

  The specific expression information 41 is included in the discourse semantics 200 as an output of the discourse structure analysis system 2, as with the flow information 21, but the usage form by the application that uses the discourse semantics 200 thereafter and the required discourse semantics are included. It may be unnecessary depending on accuracy. Therefore, the output of the specific expression information 41 is not essential, and the user may be able to select whether or not to output. Details of the specific expression analysis rule 42 and details of the processing in the specific expression analysis unit 40 will be described later.

  The cohesiveness analysis unit 50 determines each block based on the discourse data 100 after the preprocessing by the preprocessing unit 10, the flow information 21 output from the flow analysis unit 20, and the morpheme information 31 output from the morpheme analysis unit 30. The cohesiveness is analyzed and segmented, and the cohesiveness information 51 is output as an analysis result. That is, the analysis processing in the specific expression analysis unit 40 is premised on that at least morpheme analysis by the morpheme analysis unit 30 and flow analysis by the flow analysis unit 20 are performed. The cohesiveness information 51 holds segment ID information corresponding to each block ID in the discourse data 100, and is represented as a data structure using XML or the like, for example.

  The cohesiveness information 51 is included in the discourse semantics 200 as an output of the discourse structure analysis system 2 as in the case of the specific expression information 41. However, the usage form by the application using the discourse semantics 200 after that and the required discourse semantics are included. Depending on the accuracy, etc., it may be unnecessary. Therefore, the output of the cohesiveness information 51 is not essential, and the user may be able to select whether or not to output. Details of processing in the cohesiveness analysis unit 50 will be described later.

  The discourse reduction unit 60 deletes unnecessary expressions from the discourse data 100 based on the discourse data 100 after the preprocessing by the preprocessing unit 10 and the flow information 21 output from the flow analysis unit 20, and the meaning of the discourse is reduced. It has a function of extracting a certain part (hereinafter, the extracted discourse data 100 may be described as “simplification”) and outputting the reduction information 61 as a processing result. That is, the reduction process in the discourse reduction unit 60 is based on the premise that at least the flow analysis by the flow analysis unit 20 is performed. The reduction information 61 holds, for example, the ID of each statement extracted as a reduction from the discourse data 100, a newly configured block, and information on the ID, and is represented as a data structure using, for example, XML. The

  Similar to the flow information 21, the simplified information 61 is included in the discourse semantics 200 as an output of the discourse structure analysis system 1. However, the usage information by the application that uses the discourse semantics 200 after that and the required discourse semantics 200 are included. Depending on the accuracy, it may not be necessary. Therefore, the output of the reduction information 61 is not essential, and the user may be able to select whether or not to output. In the present embodiment, since the reduction information 61 is used in the discourse breakdown calculation system to be described later, the output of the reduction information 61 is essential. Details of the processing in the discourse reduction unit 60 will be described later.

  As described above, the discourse semantics 200 output as the analysis result by the discourse structure analysis system 2 includes the flow information 21, the morpheme information 31, the unique expression information 41, the cohesiveness information 51, and the reduction information 61. . At this time, a set of these data may simply be handled as the discourse semantics 200, or these data may be combined into one data structure that uses XML, for example.

[System configuration (Discourse breakdown calculation system)]
FIG. 1 is a block diagram showing an outline of a configuration example of a discourse breakdown calculation system according to an embodiment of the present invention. Similar to the discourse structure analysis system 2, the discourse breakdown calculation system 1 is configured by a computer system such as a server or a PC. For example, the dissimilarity calculation unit 70 implemented by a software program and the discourse breakdown calculation unit 80 And a table or database of the talk script 72. With the above configuration, the discourse data 101 in which the content of the discourse is converted into text and the discourse semantics 200, which is the analysis result of the discourse structure for the discourse data 101, are input, and the breakdown of the response contents for the discourse data 101 is totaled for each operator. The discourse breakdown information 300 is calculated and output.

  Note that the discourse data 101 is equivalent to that preprocessed by the preprocessing unit 10 of the discourse structure analysis system 2 described above, that is, for example, continuous statements for each speaker are collected as blocks. A sequence number (ID) indicating an order is assigned to each statement, and the speaker data of each block is set. Also, the discourse semantics 200 is output as a result of analysis of the discourse structure (flow, proper expression, cohesiveness, morpheme, and discourse reduction) for the discourse data 101 by, for example, the discourse structure analysis system 2 described above. It is a thing.

  In the present embodiment, the analysis result of the discourse structure output by the discourse structure analysis system 2 is input to the discourse breakdown calculation system 1. For example, these systems are configured on one computer system. The processing from the discourse structure analysis processing to the generation of the discourse breakdown information 300 may be executed continuously.

  Based on the discourse data 101 and the morpheme information 31 and the reduction information 61 of the discourse semantics 200, the similarity calculation unit 70 of the discourse breakdown calculation system 1 sets each block in the talk script 72 for each block of the reduced discourse data 101. It has a function of calculating the similarity in terms of expression with the talk script, obtaining the most similar talk script, and outputting the similarity information 71. The similarity information 71 holds the ID of each block, the information of the talk script most similar to this, and the information of the similarity.

  The discourse breakdown calculation unit 80 includes information about how much the operator responds to the talk script based on the discourse data 101, the flow information 21 of the discourse semantics 200, and the similarity information 71. The breakdown information is calculated and output as discourse breakdown information 300. Details of processing in the similarity calculation unit 70 and the discourse breakdown calculation unit 80 will be described later.

[Flow analysis]
Below, the detail of the process of the flow analysis by the flow analysis part 20 of the discourse structure analysis system 2 shown in FIG. 2 is demonstrated. FIG. 4 is a diagram illustrating an example of a flow and context in the discourse data 100. In the present embodiment, a context is set for each block of discourse. In the example of FIG. 4, the context of the first block (speaker: operator) is “start”, and the next block (speaker: customer) and the next block (speaker: operator) are both “declaration”. It has become.

  As shown in the example of FIG. 4, in this embodiment, in principle, the flow set in the last statement of the block immediately before a different speaker is set as the context of the target block. Since there is no immediately preceding block for the first block of the discourse data 100, “start” is always set as the context. By determining the context in this way, it is possible to analyze the flow along the discourse context even if the statements have the same wording.

  FIG. 5 is a diagram illustrating an example of the flow analysis rule 22 of the discourse structure analysis system 2. The flow analysis rule 22 is held in a file, database, or the like, and includes an expression for determining a flow in a statement (hereinafter sometimes referred to as “flow expression”) and an expression that matches the flow expression. It is a table that holds a list designated in advance with respect to correspondence with a flow pattern (hereinafter, sometimes referred to as “flow definition”) set in accordance with the context of a statement it has.

  As shown in the example of FIG. 5, the flow expression is specified by a regular expression of a character string (for example, “/Yes./” or “/ ?? /”) that serves as a clue for determining the flow. Accordingly, the flow analysis unit 20 or the discourse structure analysis system 2 needs to be able to perform processing necessary for regular expressions by introducing, for example, an existing regular expression library.

  Note that the character string specified as the flow expression should preferably specify, for example, a greeting at the beginning of a sentence, a reply, a particle, a conjunction, a sentence ending expression, etc. Specify in descending order. As for the priority, for example, the complexity of the regular expression may be calculated according to a predetermined condition, and the priority may be set in descending order of the complexity. Note that the flow expression pattern to be specified may differ depending on the type of discourse targeted by the discourse structure analysis system 2 (for example, exchange at a call center or regular meeting in a company).

  For each flow expression, for example, a flow definition is specified by a label “flow”. In the flow definition, set “greeting” as the flow if the context is “start” in the statement that matches the context of the target statement and the flow that is set in the statement (for example, “/Yes./”) Specify). A plurality of flow definitions by the label “flow” can be specified according to the type of context assumed in each flow expression.

  In addition, for example, a default flow (for example, a flow expression of “/Yes./”, which is a flow to be set when the context of the target statement does not correspond to any “flow” label by the label “default”. In the statement that matches, if the context of the statement does not match any of the “flow” labels, the flow is set to “contrast”. Note that one default flow is always set in each flow expression.

  For the flow definition, it is desirable to specify a main pattern as a general discourse flow pattern according to the type of discourse targeted by the discourse structure analysis system 2. For example, in the exchange at the call center, the flow definition can be specified based on a pattern such as “apology” if the customer makes a “complaint”.

  In the flow analysis rule 22, as described above, the flow expression to be specified and the content of the corresponding flow definition may differ depending on the type of discourse targeted by the discourse structure analysis system 2. A plurality of flow analysis rules 22 to be applied can be held so that the user can be switched, so that it is possible to flexibly cope with the characteristics of the discourse.

  As described above, in this embodiment, the part of speech is not specified in the flow analysis rule 22. That is, since the flow analysis unit 20 does not determine the part of speech, it is not necessary to perform the morpheme analysis by the morpheme analysis unit 30 in advance in the analysis process in the flow analysis unit 20. In order to improve accuracy, the morpheme information 31 is used, for example, a regular expression in the flow expression of the flow analysis rule 22 is applied to a morpheme sequence (including part-of-speech information) of the morpheme information 31 as described later. It is also possible to perform matching under more detailed conditions as applied.

  FIG. 6 is a flowchart showing an example of the flow of flow analysis processing in the flow analysis unit 20 of the discourse structure analysis system 2. When starting the flow analysis process, the flow analysis unit 20 first starts a loop process that repeats the process for each statement of the discourse data 100 after the preprocessing by the preprocessing unit 10 (S101). When loop processing for each statement is started, loop processing for repeating processing for each flow expression of the flow analysis rule 22 is started (S102).

  When the loop processing for each flow expression is started, it is determined whether or not the target statement matches the target flow expression (S103). If there is no match, the process proceeds to the next flow expression process (S108, S102). On the other hand, if there is a match, first, for the target statement, the flow set in the last statement in the block immediately before the block to which the statement belongs is set as a context (S104). At this time, as shown in FIG. 4, when the block to which the statement belongs is the first block of the discourse data 100, “start” is set in the context.

  Next, with respect to the context set in step S104, it is determined whether or not there is a context that matches the context specified by the “flow” label of the flow definition in the target flow expression (S105). If there is a match, the flow corresponding to the context in the matching flow definition is set as the flow of the target statement (S106). On the other hand, if there is no match, the default flow is set as the flow of the target statement (S107). Thereafter, the loop processing for each flow expression is exited.

  When the loop processing for each flow expression is completed, it is determined whether or not a flow is set in the target statement (S109). If the flow is set, nothing is performed. If the flow is not set, the “statement” flow is set in the target statement (S110), and the process proceeds to the next statement (S111). , S101). When all the loop processes for each statement are completed, the ID of each statement and the flow information corresponding thereto are output as the flow information 21 and the flow analysis process is terminated.

  In this way, flow analysis that considers spoken language is used to match each statement, and by setting an appropriate flow according to the context of the matched statement, flow analysis that is less susceptible to spoken language and noise for the entire discourse data 100 is performed. It can be carried out.

[Specific expression analysis]
Hereinafter, the details of the specific expression analysis processing by the specific expression analysis unit 40 of the discourse structure analysis system 2 shown in FIG. 2 will be described. FIG. 7 is a diagram showing an example of the specific expression analysis rule 42 of the discourse structure analysis system 2. The specific expression analysis rule 42 is held in a file, a database, or the like, and a regular expression for specifying a specific expression in a statement and a rule for specifying and extracting a specific expression from a portion that matches the regular expression It is a table that holds a list designated in advance for correspondence with (hereinafter may be described as “specific expression rules”).

  As shown in the example of FIG. 7, the specific expression in the statement is specified by a regular expression. Therefore, the specific expression analysis unit 40 or the discourse structure analysis system 2 needs to be able to perform processing necessary for regular expressions by introducing, for example, an existing regular expression library. In the specific expression analysis, it is necessary to determine not only the regular expression for the normal character string but also the expression of the morpheme string of each statement held in the morpheme information 31. It is expanded so that the information can be judged.

  For example, in the example of FIG. 7, the range of the morpheme string is specified by “()” in each regular expression, and this part is matched with the information of the morpheme string instead of the character string. In the first regular expression example, the part of “(<f: organization> +)” indicates one or more morpheme sequences analyzed that the part of speech (f) is “organization”. (?: Of)? "Indicates a part of speech of 0 or more and 1 or less analyzed that the surface shape is" no ". The part of “(<f: person name, surname>)” indicates one morpheme string that has been analyzed with the part of speech being “person name, surname”. This regular expression matches expressions such as “AAA Shoji Suzuki” and “AAA Shoji Suzuki”. Note that the representation of the morpheme string may vary depending on the output specifications of the morpheme analysis engine or the like in the morpheme analysis unit 30.

  At this time, in the corresponding specific expression rule, the first element in the morpheme string (the morpheme string whose part of speech is “organization”, for example, “AAA Shoji”) is specified as a specific expression having the attribute “company name”. The second element in the column (the morpheme string whose part of speech is “person name, surname”, for example “Suzuki”) is specified to be specified as a specific expression having the attribute “person name”. It should be noted that a portion (for example, “no”) enclosed by “(?:)” Is treated as not included in the element of the target morpheme string.

  In the example of the second regular expression of FIG. 7, the part of speech is specified by a normal regular expression without particular consideration, and matches the expression of a telephone number such as “03-1234-5678”. At this time, in the corresponding specific expression rule, the first element in the morpheme string (in this example, there is only one element such as “03-1234-5678”) is defined as a specific expression having the attribute “phone number”. It is specified to be specified.

  It is desirable that the character string of the unique expression specified by the regular expression is, for example, a proper noun that individually characterizes the content of the discourse or a character string made up of numerical values. In addition, for example, a character string composed of expressions (for example, complaint expression or gratitude expression) indicating the emotion of a speaker (customer, etc.) is specified and specified as a specific expression having a corresponding emotion attribute. May be. In addition, as a result of analysis, a certain character string may be included in a plurality of unique expressions.

  In the specific expression analysis rule 42, as described above, a list of correspondence between the regular expression for specifying the specific expression and the specific expression rule is specified in advance. As with the flow analysis rule 22, the normal expression to be specified is specified. For the expression and the corresponding specific expression rule, different ones are set according to the type of discourse targeted by the discourse structure analysis system 2, or the user switches the specific expression analysis rule 42 to be applied according to the type of discourse. It is possible to respond flexibly according to the characteristics of discourse.

  FIG. 8 is a flowchart showing an example of the flow of the specific expression analysis process in the specific expression analysis unit 40 of the discourse structure analysis system 2. When the specific expression analysis unit 40 starts the specific expression analysis process, first, the specific expression analysis unit 40 starts a loop process that repeats the process for each statement of the discourse data 100 (S201). When the loop processing for each statement is started, loop processing for repeating the processing for each regular expression for specifying the specific expression of the specific expression analysis rule 42 is started (S202).

  When the loop processing for each regular expression is started, it is determined whether or not each morpheme sequence in the morpheme information 31 corresponding to the target statement matches the target regular expression (S203). If there is no match, the processing proceeds to the next regular expression (S205, S202). On the other hand, if a match is found, a specific expression is identified and extracted by applying a specific expression rule corresponding to the target regular expression (S204), and the process proceeds to the next regular expression process (S205, S202).

  When all the loop processes for each regular expression are completed, the process proceeds to the next statement (S206, S201). When all the loop processing for each statement is completed, the ID of each statement and the information of the list of specific expressions corresponding thereto are output as the specific expression information 41, and the specific expression analysis process is ended.

  In this way, by performing matching with a regular expression considering the part of speech of the morpheme string and extracting the morpheme string, it is possible to perform a specific expression analysis considering a grammar peculiar to spoken words such as omission of a particle.

[Cohesion analysis]
In the following, details of cohesiveness analysis processing by the cohesiveness analysis unit 50 of the discourse structure analysis system 2 shown in FIG. 2 will be described. FIG. 9 is a diagram showing an example of cohesiveness analysis in the discourse structure analysis system 2. In the cohesiveness analysis process, first, as shown in the upper diagram, all the blocks in the discourse data 100 are set as one segment.

  Thereafter, segment division processing is performed first. In the dividing process, it is determined for each statement that a flow of a specific type is set in the flow information 21, and a segment is divided in a block to which the statement belongs. Here, it is desirable that the type of the flow to be identified is a flow that is estimated to appear when the discontinuity or content changes.

  For example, in the case of multi-person discourse such as exchanges at a call center, it is often the case that the discourse progresses depending on the response of the question and the response to it, so the content is often switched in the “question” statement of the flow. It is done. Therefore, in the present embodiment, the statement of the flow “question” (hereinafter sometimes referred to as “question statement”) is determined based on the flow information 21, and as shown in the middle diagram of FIG. Each time a question statement appears, the segment shall be split.

  Thereafter, segment integration processing is performed. In the integration process, when there is a question statement corresponding to a predetermined integration condition for each segment, the segment having the question statement is integrated with the immediately preceding segment. Here, the integration condition is designation of an expression that causes the segment having the question statement to be inferred that the content continues from the previous segment (highly cohesive with the previous segment).

  Examples of the integration condition include a case where the question statement has a word (proprietary noun) common to the immediately previous question statement or a case where a directive is included. A segment having a question statement corresponding to these integration conditions is determined to be highly cohesive with the previous segment, and is integrated with the immediately preceding segment as shown in the lower diagram of FIG. It should be noted that the integration process is repeated until there are no segments that can be integrated.

  FIG. 10 is a flowchart showing an example of the flow of cohesiveness analysis processing in the cohesiveness analyzing unit 50 of the discourse structure analyzing system 2. When the cohesiveness analysis unit 50 starts cohesiveness analysis processing, first, all the blocks of the discourse data 100 are set as one segment (S301). Specifically, for example, the initial value of the segment ID is assigned to all the blocks. After that, as a segment dividing process, a loop process for repeating the process for each block of the discourse data 100 is started (S302).

  When the loop processing for each block is started, it is determined whether or not a question statement is included in the target block (whether or not the flow has a “question” statement) (S303). If a question statement is included, the segment is divided between the target block and the immediately preceding block (S304). Specifically, for example, the next segment ID is assigned to all blocks after the target block. Thereafter, the processing moves to the next block (S305, S302).

  When the loop processing (division processing) for each block is completed, next, loop processing that repeats the processing for each segment divided by the above-described division processing is started as segment combination processing (S306). When the loop processing for each segment is started, it is determined whether or not there is a question statement corresponding to the integration condition in the target segment (S307).

  If there is a question statement corresponding to the integration condition, the target segment and the previous segment are integrated (S308). Specifically, for example, the integration is performed by updating the segment ID of each block included in the target segment to the segment ID of the immediately preceding segment. At this time, the segment ID of each block included in each subsequent segment may be sequentially incremented. Thereafter, the process proceeds to the next segment (S309, S306). When the loop processing (integration processing) for each segment is completed, information on a list of segment IDs corresponding to the ID of each block (including each block) is output as the cohesiveness information 51 for cohesiveness analysis. End the process.

  Thereby, each statement in the discourse data 100 can be segmented based on the flow information 21 obtained by the flow analysis, and can be used as useful information for various applications using the discourse semantics 200.

[Discussion reduction]
In the following, the details of the discourse reduction process by the discourse reduction unit 60 of the discourse structure analysis system 2 shown in FIG. 2 will be described. FIG. 11 is a diagram illustrating an example of discourse reduction. In the discourse reduction process, unnecessary expressions such as compliments and greetings are deleted from the discourse data 100 based on the flow information of each statement analyzed by the flow analysis unit 20, and other parts are extracted as essential parts in the discourse. To do.

  As shown in the example of FIG. 11, for the discourse data 100 for which the upper flow analysis has been performed, statements in which the flow is determined to be “greeting”, “conflict”, “standard” are deleted as unnecessary expressions, The blocks shown by the same speaker can be connected and reconstructed into one block, so that the reduction shown in the lower part can be obtained. In this embodiment, the flow that is judged to be unnecessary expression is defined as “greeting”, “conformity”, and “standard form (a flow that shows a typical expression that is used regardless of the content of the discourse)”. However, the definition of the flow that is determined as an unnecessary expression may be appropriately changed according to the type of discourse.

  FIG. 12 is a flowchart showing an example of the flow of discourse reduction processing in the discourse reduction unit 60 of the discourse structure analysis system 2. When the discourse reduction unit 60 starts the discourse reduction process, it starts a loop process that repeats the process for each statement of the discourse data 100 (S401). When the loop processing for each statement is started, the flow information 21 is referred to for the target statement, and it is determined whether the flow corresponds to any one of “greeting”, “conflict”, and “standard” (S402).

  If the flow does not correspond to any of them, the target statement is not an unnecessary expression, and a reduction flag is set for the target statement to identify this information (S403). In this embodiment, a reduction flag is set for a statement that is not an unnecessary expression. Conversely, when the target statement corresponds to any flow in step S402, an unnecessary expression flag is set as an unnecessary expression. You may make it set. Thereafter, the processing proceeds to the next statement (S404, S401).

  When the loop processing for each statement is completed, the block is reconfigured by a statement in which the reduction flag is set (the remaining statements from which unnecessary expressions are deleted) (S405). Here, when the blocks of the same speaker are consecutive due to the deletion of unnecessary expressions, these are combined into one block (for example, the last block in the discourse data 100 in the lower part of FIG. 11). . At this time, for example, when the combined block becomes longer than a predetermined length, it may be divided into a plurality of blocks having appropriate lengths. After the block is reconfigured, the block ID may be renumbered. Thereafter, the ID of each statement in which the reduction flag is set and the information of the reconfigured block are output as the reduction information 61, and the discourse reduction process is terminated.

  As a result, the content of the discourse data 100 can be compressed as faithfully as possible without leaking essential parts, and the amount of data can be reduced. Analysis efficiency and utilization in various applications that use the discourse semantics 200 thereafter. Efficiency can be improved.

  Note that the flow information 21, the unique expression information 41, the cohesiveness information 51, and the reduction information 61 included in the discourse semantics 200 output by the above-described flow analysis, specific expression analysis, cohesiveness analysis, and discourse reduction processing are as follows: The discourse structure analysis system 2 may be provided with an interface that allows the user to refer to the contents and make corrections as appropriate.

[Discourse breakdown calculation]
Hereinafter, the details of the process of calculating the discourse breakdown information by the similarity calculation unit 70 and the discourse breakdown calculation unit 80 of the discourse breakdown calculation system 1 shown in FIG. 1 will be described. FIG. 13 is a diagram illustrating an example of similarity calculation in the discourse breakdown calculation system 1. The upper left part of FIG. 13 shows an example of the contents of each block included in the discourse data 101, and the upper right part shows an example of the contents of each talk script included in the talk script 72.

  As shown in the example of FIG. 13, each script in the talk script 72 is a script for each series of utterances defined for each operation that the operator supports (for example, “cancellation” or “address change” in insurance operations). An ID is assigned, and text indicating the content (content to be uttered) is defined. Also, by defining the next script ID, it is possible to define a business flow consisting of a series of talk scripts. It is also possible to specify a condition for branching the definition of the next script according to the response content or situation of the customer.

  The similarity calculation unit 70 of the discourse breakdown calculation system 1 calculates the similarity in expression between the text of each block of the discourse data 101 and the text of each talk script of the talk script 72, and starts from the text with the highest similarity. Identify the talk script. When this similarity is higher than a predetermined threshold, the speaker (in this embodiment, the operator) speaks according to the specified talk script (the highest similarity talk script) in the target block. Can be estimated. The information of the talk script having the highest similarity with the target block and the value of the similarity are output to the similarity information 71 as shown in the lower part of FIG.

  Even if the operator utters according to one talk script, if a statement such as “greeting” or “consideration” is included in between, the block in the discourse data 101 is divided by these and spans multiple blocks. It becomes a state. In such a situation, it is not possible to properly compare with the text of each talk script. Therefore, in this embodiment, it is assumed that the information of each block is acquired from the result of reducing the discourse data 101 based on the reduction information 61 included in the discourse semantics 200. That is, unnecessary expressions such as “greeting” and “consideration” are deleted from the discourse data 101, and further, information on each block is acquired from a block reconstructed into a single block by connecting the same speaker. By using the blocks obtained here, it is possible to make a more effective comparison by combining the granularity of both texts.

  The means for calculating the similarity in expression between the text of each block and the text of each talk script in the simplified discourse data 101 is not particularly limited, and various means can be used. For example, as a commonly used technique, the length of a so-called longest common subsequence is calculated by a LCS (Longest Common Subsequence) algorithm for a sequence of words included in both texts, and the similarity is obtained by normalizing this. be able to.

  The content of the discourse for each speaker (operator) based on the similarity information 71 that is the result of calculating similarity information for each block for each discourse included in the discourse data 101 and the flow information 21 in the discourse semantics 200. The discourse breakdown information 300 can be obtained. FIG. 14 is a diagram showing a specific example of the discourse breakdown information 300. In the example of FIG. 14, the content of the discourse breakdown information 300 is displayed on the terminal screen in a table format.

  FIG. 14A shows information obtained by tabulating the breakdown of all discourses (responses) every month for each operator. Here, for example, as the breakdown of the response count and the average response time, and the content of the response contents, the rate of performing “greeting” and “conversation” (%), utterances that match any of the talk scripts in the talk script 72 were made. The percentage and the percentage of utterances that do not correspond to any of the above are tabulated and displayed. Similarly, FIG. 14B shows the detailed information of FIG. 14A, summing up and displaying the breakdown of the response time and response contents for each talk (response) for each operator.

  In FIG. 14, for example, when the ratio of conflicts is large, it can be determined that the customer is often listening to the utterance. If the ratio of utterances that match the talk script and the ratio of other utterances is high, the utterance that matches the talk script is high. It can be evaluated that the talk script is valid. On the other hand, if the ratio is too high, it can be evaluated that the customer tends to give an impression that the reception is clerical.

  On the other hand, when the ratio of other utterances is high, there may be factors such as the case where the content shown in the talk script is uttered by changing the expression without permission, or the story often deviates sideways. It can also be evaluated that there is no talk script suitable for the situation, or that the talk script is not useful. On the other hand, it may be evaluated that the operator flexibly responds to the customer's utterance flexibly.

  Therefore, it may be necessary to make a judgment by referring to the actual content of the discourse. In each table in FIG. 14, for example, each month or each response start date / time item in the leftmost column is set as a hyperlink, and the corresponding discourse content is acquired from the discourse data 101 and displayed by selecting the corresponding link. You may make it do.

  FIG. 15 is a flowchart showing an example of the flow of the discourse breakdown calculation process in the similarity calculation unit 70 and the discourse breakdown calculation unit 80 of the discourse breakdown calculation system 1. When the discourse breakdown calculation process is started, first, the similarity calculation unit 70 starts a loop process that repeats the process for each block of the discourse data 101 reduced by the reduction information 61 (S501). When the loop processing for each block is started, a word for which the similarity is calculated by the LCS algorithm is extracted from the text in the target block (S502). Here, based on the morpheme information 31, only verbs, nouns, and adjectives are extracted. For verbs and adjectives, the original form is used. As a result, it is possible to improve the accuracy by preventing the similarity from being lowered more than necessary because it does not match due to a subtle difference in wording.

  Next, a loop process for repeating the process for each talk script in the talk script 72 is started (S503). When the loop processing for each talk script is started, morpheme analysis is performed on the target talk script, and it is divided into morpheme strings to determine the part of speech (S504). At this time, as in step S502, the original form is used for verbs and adjectives. Also, only verbs, nouns, and adjectives may be extracted. For the morpheme analysis, an existing morpheme analysis engine can be used, or the morpheme analysis unit 30 of the discourse structure analysis system 2 may be used.

  Next, an expression similarity is calculated between each word in the target block acquired in step S502 and each word in the target talk script acquired in step S504 (S505). For example, as described above, the length of the longest common subsequence can be calculated by the LCS algorithm and normalized to obtain the similarity. Thereafter, the process proceeds to the next talk script process (S506, S503).

  When the loop processing for each talk script is completed, the talk script having the highest similarity calculated in step S505 is selected and output to the similarity information 71 (S507). Thereafter, the process proceeds to the next block (S508, S501). When the loop processing for each block is completed, the breakdown of the similarity information 71 is aggregated for each operator, and the discourse breakdown information 300 is calculated and output (S509), and the discourse breakdown calculation processing ends.

  When calculating the discourse breakdown information 300, for example, whether the utterance content of each block is “greeting” or “conversation” can be determined with reference to the flow information 21. Whether or not the utterance is in accordance with the talk script can be determined based on whether or not the similarity calculated in step S505 is higher than a predetermined threshold. Based on the discourse breakdown information 300 calculated in this way, for example, an output screen as shown in FIG. 14 can be displayed.

  As described above, according to the discourse breakdown calculation system 1 according to an embodiment of the present invention, the similarity in expression between each block of the simplified discourse data 101 and each talk script in the talk script 72. By calculating the degree, it is determined which talk script the operator's utterance is most similar to. Based on this information and the flow information 21 included in the discourse semantics 200, the breakdown of the operator's discourse (whether it is a greeting or companion, whether it is an utterance that matches the talk script, etc.) can be efficiently performed with sufficient accuracy. It is possible to calculate.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

  INDUSTRIAL APPLICABILITY The present invention can be used for a discourse breakdown calculation system and a discourse breakdown calculation program that generates breakdown information about the content of a discourse from the result of analyzing the structure of discourse data obtained by speech recognition processing.

1 ... Discourse breakdown calculation system, 2 ... Discourse structure analysis system, 3 ... Speech recognition engine,
10: Pre-processing unit,
20 ... Flow analysis unit, 21 ... Flow information, 22 ... Flow analysis rule,
30 ... Morphological analyzer, 31 ... Morphological information,
40 ... proper expression analysis unit, 41 ... proper expression information, 42 ... proper expression analysis rule,
50: Cohesiveness analysis unit, 51 ... Cohesiveness information,
60 ... Discussion reduction part, 61 ... Simple information,
70 ... similarity calculation unit, 71 ... similarity information, 72 ... talk script,
80 ... Discourse breakdown calculation unit,
100, 101 ... Discourse data, 200 ... Discourse semantics, 300 ... Discourse breakdown information.

Claims (10)

With the input of discourse semantics, which is information for grasping the semantic content of the whole discourse, which is the analysis result of the discourse structure in the discourse data, and the discourse data in which the content of the discourse is converted into text by voice recognition processing, A discourse breakdown calculation system that calculates and outputs the breakdown information of a discourse by a speaker, including the usage status of reference information referred to by a speaker during a discourse,
The discourse semantics includes simplified information indicating information about blocks that are composed of one or more consecutive statements for each speaker, and statements that do not fall under unnecessary expressions in the discourse data,
Response time reference information that holds information about one or more of the reference information referred to in the discourse;
Using the block in the discourse data reduced based on the reduction information as a unit, the similarity in expression between the text included in the block and the text of each reference information is calculated, and the reference most similar to the block A similarity calculator that obtains information and outputs it as similarity information;
Based on the similarity information, when the similarity for each block in the simplified discourse data is higher than a predetermined threshold, it is determined that the block is uttered according to the corresponding reference information, A discourse breakdown calculation unit that calculates and outputs discourse breakdown information including information on the ratio of blocks that are uttered according to the reference information and blocks that are not uttered according to the reference information, for each speaker. Discourse breakdown calculation system. In the discourse breakdown calculation system according to claim 1,
The discourse semantics includes flow information including flow information indicating the semantic content of each statement in the discourse data,
The discourse breakdown calculation unit further calculates, as the information included in the discourse breakdown information, a ratio indicating that a flow set in a statement included in each block in the discourse data is a greeting or a match, respectively. Discourse breakdown calculation system as a feature. In the discourse breakdown calculation system according to claim 1 or 2,
The discourse semantics includes morpheme information including information on a morpheme string obtained by dividing each statement in the discourse data into morphemes and determining a part of speech;
Based on the morpheme information, the similarity calculation unit obtains the longest common subsequence of words between the text included in the block in the simplified discourse data and the text of each reference information. Discourse breakdown calculation system characterized by calculating similarity of In the discourse breakdown calculation system according to claim 3,
When calculating the longest common subsequence, the similarity calculation unit uses only verbs, nouns, and adjectives for text included in the simplified block of discourse data, and uses the original form for verbs and adjectives. Discourse breakdown calculation system characterized by In the discourse breakdown calculation system according to any one of claims 1 to 4,
The discourse breakdown calculation system, wherein the reference information is a talk script or FAQ for customer reception by an operator. With the input of discourse semantics, which is information for grasping the semantic content of the whole discourse, which is the analysis result of the discourse structure in the discourse data, and the discourse data in which the content of the discourse is converted into text by voice recognition processing, A discourse breakdown calculation program that causes a computer to function as a discourse breakdown calculation system that calculates and outputs the breakdown information of discourse by a speaker, including the usage status of reference information referred to by a speaker during a discourse,
The discourse semantics includes simplified information indicating information about blocks that are composed of one or more consecutive statements for each speaker, and statements that do not fall under unnecessary expressions in the discourse data,
Having reference information at the time of holding information about one or more of the reference information to be referred to at the time of discourse;
Using the block in the discourse data reduced based on the reduction information as a unit, the similarity in expression between the text included in the block and the text of each reference information is calculated, and the reference most similar to the block Similarity calculation processing for acquiring information and outputting it as similarity information;
Based on the similarity information, when the similarity for each block in the simplified discourse data is higher than a predetermined threshold, it is determined that the block is uttered according to the corresponding reference information, A discourse breakdown calculation process for calculating and outputting discourse breakdown information including information on a block spoken in accordance with the reference information and a ratio of blocks not spoken in accordance with the reference information for each speaker is performed. Discourse breakdown calculation program. In the discourse breakdown calculation program according to claim 6,
The discourse semantics includes flow information including flow information indicating the semantic content of each statement in the discourse data,
The discourse breakdown calculation process further calculates, as information included in the discourse breakdown information, a ratio indicating that a flow set in a statement included in each block in the discourse data is a greeting or a match. A featured discourse breakdown calculation program. In the discourse breakdown calculation program according to claim 6 or 7,
The discourse semantics includes morpheme information including information on a morpheme string obtained by dividing each statement in the discourse data into morphemes and determining a part of speech;
The similarity calculation processing is based on the morpheme information to obtain the longest common subsequence of words between the text included in the block in the simplified discourse data and the text of each reference information. The discourse breakdown calculation program characterized by calculating the similarity of. In the discourse breakdown calculation program according to claim 8,
When calculating the longest common subsequence, the similarity calculation process uses only verbs, nouns, and adjectives for text included in the simplified block of discourse data, and uses the original form for verbs and adjectives. A discourse breakdown calculation program characterized by In the discourse breakdown calculation program according to any one of claims 6 to 9,
The discourse breakdown calculation program characterized in that the reference information is a talk script or FAQ for customer service by an operator.

Images